> On Fri, Aug 04, 2006 at 11:58:22AM +0200, henka@xxxxxxxxxxxxx wrote: >> > On Fri, Aug 04, 2006 at 10:48:17AM +0200, henka@xxxxxxxxxxxxx wrote: >> > Two big questions: >> > >> > 1. What encoding are the two database (\l will tell you)? >> > 2. What encoding are the clients expecting? > >> I've even tried using LATIN1 (ie, explicitly setting it to latin1 using >> initdb, and then restoring the database after changing the 'utf-8' >> strings >> in the dump data to 'latin1'). This still yields the funny chars. > > Wait, so the dump is in utf-8? You shouldn't need to edit the dump, > postgresql will convert the encodings on the fly while loading. I've actually found two versions - one with UTF-8, and the other LATIN1. >> To be honest, I have no idea what the origional encoding was. > > It should be in the dump file, almost the first line. Locale is of no > interest to pg_dump, you'll have to decide how you want it. Yes: UTF-8 and the other is LATIN1 > Well, at the very least, does it go away if you type: > > set client_encoding=latin1; No it doesn't. That was one of the first things I tried after reading the docs. > Please provide more details about your setup too, your client is on > windows? The server is ...? Server: Linux debian sarge PG: 8.1.4 show all: lc_collate C lc_ctype C lc_messages C lc_monetary C lc_numeric C lc_time C client_encoding LATIN1 (or UTF-8) Clients: Windows using PuTTY (ie, for psql), and dynamic web content with PHP/Pg (on any browser).
